add url textract for linkedin

2 files changed, 58 insertions, 0 deletions

diff --git a/internal/platforms/linkedin/escapes.go b/internal/platforms/linkedin/escapes.go
index 5f803a5..21b4c56 100644
--- a/internal/platforms/linkedin/escapes.go
+++ b/internal/platforms/linkedin/escapes.go
@@ -1,6 +1,7 @@
 package linkedin
 
 import (
+	"regexp"
 	"strings"
 )
 
@@ -37,3 +38,17 @@ func escapeLinkedInText(input string) string {
 
 	return builder.String()
 }
+
+// extractURLs finds all occurrences of URLs starting with "http://" or "https://" in a given string.
+func extractURLs(input string) []string {
+	// Regular expression pattern to match URLs starting with http:// or https://
+	urlPattern := `(http://|https://)[^\s]+`
+
+	// Compile the regular expression
+	re := regexp.MustCompile(urlPattern)
+
+	// Find all matches in the input string
+	urls := re.FindAllString(input, -1)
+
+	return urls
+}
diff --git a/internal/platforms/linkedin/escapes_test.go b/internal/platforms/linkedin/escapes_test.go
index b1f9203..2074312 100644
--- a/internal/platforms/linkedin/escapes_test.go
+++ b/internal/platforms/linkedin/escapes_test.go
@@ -1,6 +1,8 @@
 package linkedin
 
 import (
+	"fmt"
+	"slices"
 	"testing"
 )
 
@@ -13,3 +15,44 @@ func TestLinkedInEscapes(t *testing.T) {
 		t.Errorf("expected '%s' but got '%s'", expected, escaped)
 	}
 }
+
+func TestLinkedInTwoURLsExtract(t *testing.T) {
+	text := `Hello world https://foo.zone
+	Hello universe http://world.universe test 123`
+
+	urls := extractURLs(text)
+	if len(urls) != 2 {
+		t.Errorf("expected 2 URLs, but got %d", len(urls))
+	}
+
+	if !slices.Contains(urls, "https://foo.zone") {
+		t.Errorf("expected 'https://foo.zone' in the URL list, but got %v", urls)
+	}
+	if !slices.Contains(urls, "http://world.universe") {
+		t.Errorf("expected 'http://world.universe' in the URL list, but got %v", urls)
+	}
+}
+
+// TODO: Use Fuzzing here!
+func TestLinkedInURLExtract(t *testing.T) {
+	urls := []string{
+		"http://foo.zone",
+		"http://foo.zone/",
+		"http://foo.zone?foo=bar",
+		"http://foo.zone/?foo=bar",
+		"http://foo.zone/?foo=bar",
+		"http://foo.zone/hurs?foo=bar",
+		"http://foo.zone?foo=bar&baz=bay",
+	}
+
+	for _, url := range urls {
+		text := fmt.Sprintf("Hello world %s Hello World", url)
+		found := extractURLs(text)
+		if len(found) != 1 {
+			t.Errorf("expected 1 URL, but got %d for text '%s'", len(found), text)
+		}
+		if found[0] != url {
+			t.Errorf("expected URL '%s', but got '%s' for text '%s'", url, found[0], text)
+		}
+	}
+}